Use LithosAI with Hermes Agent

Hermes Agent connects directly to LithosAI through the OpenAI Chat Completions API. You do not need a local proxy or protocol adapter.

This setup uses a named custom provider, so you can keep LithosAI alongside any other providers already configured in Hermes Agent.

Before you begin

Create a key on the API Keys page and confirm the model ID you want to use on the Models page.

Install Hermes Agent

Install Hermes Agent with any method on the official installation page, then confirm the installation.

hermes doctor

Add your LithosAI key

Hermes Agent reads provider keys from ~/.hermes/.env. Add your key there.

~/.hermes/.env

LITHOSAI_API_KEY=your-key

Exporting LITHOSAI_API_KEY in the shell that starts Hermes Agent also works.

Add LithosAI as a custom provider

LithosAI is a named custom provider. Merge the following into ~/.hermes/config.yaml, or open the file with hermes config edit.

~/.hermes/config.yaml

providers:
  lithosai:
    api: https://api.lithosai.cloud/v1
    key_env: LITHOSAI_API_KEY
    transport: chat_completions
    default_model: moonshotai/Kimi-K3

model:
  default: moonshotai/Kimi-K3
  provider: custom:lithosai

key_env names the environment variable that holds the key, so the credential never enters the configuration file. Replace the model ID if you selected a different model in the LithosAI console.

Hermes Agent does not read the context window from LithosAI. If the context meter or compaction looks wrong, add context_length to the lithosai entry with the value from the Models page.

Start a session

Start Hermes Agent from the project you want it to inspect. The session opens on LithosAI because model.provider is custom:lithosai.

hermes

Auxiliary tasks such as vision, web summarization, and session titles use the main chat model by default, so they run on LithosAI too.

Verify the connection

From a project that contains a README.md, ask Hermes Agent:

Read README.md and reply with only its first Markdown heading.

Hermes Agent should read the local file and return its heading. This verifies both the model response and a basic local tool call.

Troubleshooting

Symptom	Cause
`Unknown provider` or LithosAI does not appear	`model.provider` must be `custom:lithosai`, and `providers.lithosai` must be present in `~/.hermes/config.yaml`.
Authentication fails	`LITHOSAI_API_KEY` is not in `~/.hermes/.env` or the shell that started Hermes Agent.
Context usage or compaction is inaccurate	Add `context_length` to the `lithosai` provider entry.

Issues common to every agent are listed under Common issues.

Getting Started

Inference APIs

Account

Use LithosAI with Hermes Agent

Before you begin

Verify the connection

Troubleshooting

Resources

​Before you begin

​Verify the connection

​Troubleshooting

​Resources

Before you begin

Verify the connection

Troubleshooting

Resources